INFORMATION AND KNOWLEDGE MANAGEMENT IN NATURE CONSERVATION

INFORMATION AND KNOWLEDGE MANAGEMENT IN NATURE CONSERVATION

Supporting planning, decision-making, monitoring and evaluation

in wildlife management in Uganda

with a spatial management information system (MIST)

Klaus Schmitt and Kevin Sallee

2002 Deutsche Gesellschaft fürTechnische Zusammenarbeit (GTZ) GmbH

Information and Knowledge Management in Nature Conservation

i

Contents

SUMMARY..................................................................................................................................4

1 Introduction ......................................................................................................................6

2 The approach to the development of MIST.....................................................................8

3 The organisational set-up at UWA headquarters ...........................................................9

4 The management information system - MIST...............................................................11

4.1 Management information needs determine data collection, analysis and outputs ....114.2 Standardisation of data collection and outputs.........................................................114.3 Processing of all data and information relevant to management ..............................134.4 Use of geo-referenced data for planning, decision-making, monitoring and

evaluation................................................................................................................144.5 Fast and accurate data entry ...................................................................................174.6 Central information system (database maintenance and integrity)...........................194.7 Access for all users .................................................................................................234.8 User-friendly programming contributes to sustainability...........................................25

5 Management effectiveness ............................................................................................26

5.1 The management cycle ...........................................................................................265.2 Measuring management effectiveness ....................................................................28

6 Meta-database ................................................................................................................31

7 Capacity building through information and knowledge management........................32

For further information contact:Dr Klaus Schmitt [email protected] Sallee [email protected] Wildlife Authority [email protected]


ii

List of figures

Fig. 1 Location of the National Parks (NP) in Uganda...........................................................6

Fig. 2 Position of MIST within UWA at the beginning of the project.......................................9

Fig. 3 Position of the Information Management Unit within the organisation .........................9

Fig. 4 The MIST toolbar and an overview of the MIST databases.......................................14

Fig. 5 Management sectors of Murchison Falls Conservation Area and areas patrolledduring October 1998 .................................................................................................15

Fig. 6 The three logical units of MIST (database/server, libraries and interface) .................20

Fig. 7 MIST Query Wizards: A - for predefined outputs, B - for flexible output creation.......22

Fig. 8 Database Administrator Tool in MIST .......................................................................23

Fig. 9 Data and information flow and user access in the UWA Management InformationSystem .....................................................................................................................24

Fig. 10 Hierarchy of planning in UWA...................................................................................26

Fig. 11 Annual management cycle........................................................................................27

Fig. 12 Flowchart to illustrate how the evaluation assessment codes A – C are assigned ....28

Fig. 13 Relationship between the key components of the information system andknowledge management...........................................................................................32

List of tablesTable 1 Overview of data recorded by ranger patrols ............................................................18

Table 2 Framework for assessing the management effectiveness of protected areas andprotected area systems ............................................................................................26

Table 3 The assessment levels in MIST-AOP........................................................................29


iii

List of Acronyms

AOP Annual Operations Plan

CITES Convention on International Trade in Endangered Species of Wild Fauna and Flora

EIA Environmental Impact Assessment

ESB Environmental Scoping Brief

ESRI Environmental Systems Research Institute

GIS Geographic Information System

GMP General Management Plan

GPS Global Positioning System

GRID Global Resource Information Database

GTZ Deutsche Gesellschaft für Technische Zusammenarbeit (German TechnicalCooperation)

HQ Headquarters

IMU Information Management Unit

IS Information System

LC Local Council

MFCA Murchison Falls Conservation Area

MIKE Monitoring Illegal Killing of Elephants

MIST Management Information System

NASA National Aeronautics and Space Administration

NP National Park

PA Protected Area

PC Personal Computer

SQL Structured Query Language

USh Uganda Shilling

UNEP United Nations Environment Programme

UWA Uganda Wildlife Authority


Page 4

SUMMARYThe GTZ project ‘Advisory Services to Uganda Wildlife Authority (UWA)’ has been supportingUWA with the development of a spatial Management Information System (MIST) to providemanagers and planners with up-to-date information for planning, decision-making andevaluation. Important aspects taken into consideration during the development of the systemwere the organisational set-up of UWA, behavioural principles and technical factors.

MIST has been placed in an Information Management Unit which reports directly to theExecutive Director of UWA to guarantee access for all users, horizontal information flow, dataintegrity and use of meta-data.

The intended users of MIST were involved in the design of the information system right from thestart. Their information needs determine what data are collected, the methods of analysis andthe types of outputs.

MIST has been developed as a user-friendly client/server application programme which is easyto maintain and up-date, needing only a computer specialist who can manage a computernetwork. This avoids the problems associated with a system which can only be used byspecialised GIS (Geographic Information System) experts and is thus unsustainable.

All users in UWA Headquarters have easy access to the central database from their workspacethrough a local area network. The users in the protected areas (PAs) have access to MISTthrough stand alone computers in the PAs. The databases from the PAs are synchronised withthe central MIST database through automatic database replication using digital data transfer.For PAs without a telephone link to UWA Headquarters data transfer and database replication isdone using zip-disks.

MIST provides information generated by the processing of data and also by giving access tofiles which are kept in central digital archives on the UWA server. Only data which can beprocessed into information which is relevant and useful for managers and planners arecollected, stored and processed in MIST. Fast and accurate data entry is guaranteed throughdata download and the use of look-up tables for data entry. Standardised data sheets havebeen developed for the recording of (1) ecological data and data on illegal activities by rangersand air patrols, (2) visitor data at park gates and (3) data on resource harvest by localcommunities. Data collection by rangers on law-enforcement patrols is simple, fast, designed tominimise data collection mistakes and provides raw data for a wide range of outputs.

All spatial data collected are geo-referenced using Global Positioning System (GPS). Thisenables production of the following outputs: (1) indices for monitoring of wildlife populations,illegal activities and resource harvest by local communities without the need for expensivebaseline data, (2) distribution maps for planning and monitoring, (3) baseline information forpatrol deployment planning, (4) information on wildlife population structure, (5) performanceindicators to monitor and evaluate the implementation of annual operations plans and(6) reports for collaborating institutions etc.

Annual operations planning and monitoring and evaluation of implementation of plans are alsohandled in MIST.

The programme consists of three logical units: a database and database server software,application processing libraries and a graphical user interface. Borland Delphi was used todevelop the user interface and libraries for MIST. Borland's Interbase, version 6.0, SQL(Structured Query Language) database server handles data storage and processing. Thespatial display is handled by ESRI MapObjects using shape files. The programme can becompletely controlled by clicking on map items, page tabs, list entries and buttons of the MIST


Page 5

query wizards. A tool to refine the SQL WHERE clause created by a query wizard and a visualSQL statement wizard provide flexible ways to create any query not included in the custom-made wizards. Most of the maintenance of the MIST database can be done using a visual MISTDatabase Administrator Tool.

The programming of MIST contributes to sustainability through (1) a user-friendly interface (i.e.managers do not have to learn new and complicated software), (2) the use of programmelibraries, each dealing with a single logical task which can be up-dated on an as-needed basisand (3) the possibility to up-date parts of the programme using SQL procedures and triggersinstead of having to re-code the MIST user-interface or the Windows libraries.

MIST contributes to institutionalising information and knowledge management by providing allstaff in UWA with easy access to the central database, the meta-database and all documentsand manuals of relevance to Uganda Wildlife Authority. It also contributes to creating a cultureof knowledge exchange.

MIST is a custom-made, easy to use, flexible and powerful tool to improve management throughaccess to up-to-date information, which can also be used to measure the effectiveness ofmanagement.


Page 6

1 INTRODUCTION

The Uganda Wildlife Authority (UWA) manages 10 National Parks (see figure 1) and 13 WildlifeReserves in Uganda. In addition to these wildlife protected areas (PAs), UWA is responsible for7 Wildlife Sanctuaries and provides guidance for 5 Community Wildlife Areas.

Fig. 1 Location of the National Parks (NP) in Uganda

As a result of years of civil war and insecurity, management of protected areas in Uganda waslargely reduced to anti-poaching and policing measures. Following the merger in 1996 ofUganda National Parks and the Game Department to form the Uganda Wildlife Authority, GTZ(German Technical Cooperation) in June 1997 started the project Advisory Services to UgandaWildlife Authority, to contribute to institutional and management capacity building in UWA. Theproject purpose is:

• to provide accurate, up-to-date information on ecological, social and economic dimensionsof wildlife conservation for decision-making.

Mount Elgon NP

Kidepo Valley NP

Murchison Falls NP

Semuliki NP

Rwenzori Mountains NPKibale NP

Queen Elizabeth NP

BwindiImpenetrable NP

Mgahinga Gorilla NP

LakeMburo NP

Lake KyoggaLake Albert

Lake Edward

Lake Victoria 0 100 km

Uganda’s National Parks


Page 7

The project purpose can best be achieved through the establishment of an information system(IS), which provides an effective way of storing and processing data into information forplanning, decision-making and evaluation. Timely dissemination of this information to managers,in an appropriate form, must be an integral component of such an IS.

The terms data and information are often used interchangeably. However, in the context ofinformation management they have clearly distinct meanings:

• data are facts which have been recorded and• information is processed data.

It is important to note that data must be processed in such a way that the resulting informationcan be understood and is useful to the recipient.

The GTZ project has been supporting Uganda Wildlife Authority with the development of aspatial Management Information System (MIST).

An information system is "a system to convert data from internal and external sources intoinformation and to communicate that information, in an appropriate form, to managers at alllevels in all functions to enable them to make timely and effective decisions for planning,directing and controlling the activities for which they are responsible."

(Lucey, T 1997: Management Information Systems. 8th edition, Letts. London. 282 pp.)


Page 8

2 THE APPROACH TO THE DEVELOPMENT OF MIST

For the development and the implementation of an information system which provides up-to-date information to managers in all protected areas and at all levels of decision-making, keyfactors to be considered were:

• organisational set-up• technical factors and• behavioural principles

In developing MIST a bottom-up approach was used to ensure that the users of the IS wereinvolved in its design right from the start. The system has been tailor-made for the specificneeds of the different users who therefore can identify themselves with the system.

The process of system development began with a pilot study conducted in one protected area1,the aim of which was to collect data, set standards and develop the computer programmefollowing the steps outlined below.

Building on what has been learnt during the pilot study these steps were carried out in eachprotected area, for each department in the Uganda Wildlife Authority Headquarters, for theBoard of Trustees and for the Ministry of Tourism, Trade and Industry.

1 Murchison Falls Conservation Area - MFCA, in north-west Uganda.

The following steps were taken in the development of the information system:

1. Assessment of information needs2. Design of suitable standardised data collection and analysis procedures3. Development of the computer system and application programme4. Use of the Information System


Page 9

3 THE ORGANISATIONAL SET-UP AT UWA HEADQUARTERS

When the project started, MIST was placed in the Monitoring and Research Unit of theDepartment of Planning, Monitoring and Research (Fig. 2).

Information management anddissemination to alldepartments in UWAHeadquarters (HQ) and allprotected areas were taskswhich could not be doneeffectively within thisdepartmental structure. Toaddress this problem, UgandaWildlife Authority hasestablished an InformationManagement Unit (IMU) whichreports directly to the ExecutiveDirector (Fig. 3). MIST is nowlocated within this IMU.

Fig. 2 Position of MIST within UWA at the beginning of theproject

Fig. 3 Position of the Information Management Unit within the organisation

The IMU’s function is to create an information network linking the departments of UWAHeadquarters and the protected areas. It is responsible for ensuring that duplication of work andrecording of identical data do not happen. It is also responsible for overseeing that individualdepartments and protected areas do not use or produce different data versions of the sametopic. This guarantees the integrity of data created in UWA. To support this data warehousingfunction the IMU maintains a meta-database which serves as a reference for all users (forfurther details see section 6).

Director PlanningMonitoring and

Research

Director FieldOperations

Director TourismDevelopmentand Marketing

ExecutiveDirector

Internal AuditorIMU

Director HumanResources andAdministration

DirectorFinancialServices

Director PlanningMonitoring and

Research

ExecutiveDirector

Monitoring andResearch

Co-ordinator

MIST

Planning andEIA Co-ordinator

Director FieldOperations

DirectorTourism


Page 10

The role of MIST, as the centre of the information network, is:

• to serve as an UWA in-house decision-support tool,• to provide information for planning and evaluation,• to act as a central data depository for incoming multi-sectoral data, and• to provide data and information to other institutions (e.g. in the context of the Convention of

Biological Diversity and CITES, the Convention on International Trade in EndangeredSpecies of Wild Fauna and Flora)

MIST has been placed in an Information Management Unit which reports directly to theExecutive Director to guarantee data integrity, use of meta-data, access for all users andhorizontal information flow.


Page 11

4 THE MANAGEMENT INFORMATION SYSTEM - MIST

Key points for the development of MIST, which provides UWA management, at all levels ofdecision-making, with up-to-date information, are:

4.1 Management information needs determine data collection, analysis andoutputs

At the beginning of the pilot study meetings were held with park managers to identify the kind ofinformation they needed for day to day decision making. The information needs determinedwhat data were collected, the methods of analysis and the types of output – the computersystem was then programmed to provide the required information. This demand-drivenapproach ensures that MIST is a tool which does what the users want.

The management of MFCA defined their information needs, based on ecological data and dataabout illegal activities, as follows:

• wild animals: what species; sex, age distribution, group size; location of sighting and habitat• vegetation: vegetation type; burning and amount of dead matter• utilisation indication: poacher/camps; felled trees; pit saws; charcoal burning and

encroachment• others: salt/mineral licks; water holes and patrol coverage

4.2 Standardisation of data collection and outputs

No up-to-date field data2 on wildlife and illegal activities were available when the project started.This provided the opportunity to establish data collection standards which could be applied in allPAs. With such standards in place long familiarisation periods and retraining are no longernecessary when staff are transferred. Furthermore, standardised procedures also lessen thenegative effect if individuals leave the organisation, because enough people with knowledgeremain to train new staff.

2 Wildlife population estimates based on systematic reconnaissance flights for all protected areas werecarried out in 1995/96.

• Management information needs determine data collection, analysis and outputs• Standardisation of data collection and outputs• Processing of all data relevant to management• Use of geo-referenced data for planning, decision-making, monitoring and evaluation• Fast and accurate data entry• Central information system (database maintenance and integrity)• Access for all users• User-friendly programming contributes to sustainability


Page 12

At present UWA has neither the manpower nor the money to carry out systematic monitoringprogrammes and thus decided to use ranger patrols for collection of monitoring data3. However,systematic data collection for wildlife monitoring by UWA staff has started for selected specieswhich can be monitored easily and do not require time- and resource-consuming methods.Examples are the regular recording of fish eagle and hippopotamus numbers along waterways.

Data collection

Important criteria which were taken into consideration for the development and design of datasheets were:• Only data which could be processed into information useful to managers were to be

collected.• Data recording needed to be easy and fast.• Handling of data sheets in the field needed to be easy (i.e. only one page and not too many

columns and tick boxes) without sacrificing details required by managers.• Manual data entry was replaced with data import wherever possible.• It needed to be possible to access the data in different ways for various purposes/ outputs.

Based on the information needs of managers and planners, data sheets standardised for usethroughout the country, were developed for ranger patrols4. Ranger patrols record observationson key wildlife species (total numbers, sex distribution of adults and number of young) and onillegal activities.

A standardised procedure for resource off-take monitoring by local communities is currentlybeing tested in the field (for details see 4.4).

Training manuals and comprehensive instruction leaflets have been developed and trainershave been trained.

Outputs

Outputs from MIST can be based on

• routine information requests and• specific information requests.

For routine information requests, MIST provides managers with standardised monthly (quarterlyand annual) outputs. For example, standard monthly outputs produced by MIST for MurchisonFalls Conservation Area from data collected by ranger law-enforcement patrols include maps ofobservations and patrolled areas and reports on patrol performance indicators, key mammalspecies and illegal activities.

In addition to maps and tables MIST can produce graphics, charts and diagrams to presentinformation from any kind of request in an easily understandable way.

3 A research programme is underway in Murchison Falls National Park to examine whether data collectedopportunistically by ranger law-enforcement patrols can be used to obtain precise indices for monitoringtrends in wildlife populations and illegal activities in discrete and relatively small areas such asmanagement sectors. Schmitt, K., Jaeckel, W. and Atukunda-Muhimbura, A. (in press): An informationsystem for wildlife monitoring in Uganda. - In: Proceedings of the Second International WildlifeManagement Congress: 157-160.4 In addition, standards for routine air surveillance data sheets were developed in MFCA.


Page 13

4.3 Processing of all data and information relevant to management

The system has been designed, both in terms of software and hardware, to handle largeamounts of data in an efficient manner.

Data collection and data entry are expensive and it is therefore important that only data whichcan be processed into information which is relevant and useful for managers and planners arecollected and stored.

Data used in MIST come from a wide spectrum of sources. They include data collected byrangers and gate clerks, data from air surveillance, systematic and total wildlife counts from theair and along roads and rivers, data collected by local communities and by researchers. Inaddition MIST uses data and information from the UWA accounting software and FinancialManagement System5.

MIST also handles data entry and management of the annual operations planning process andthe evaluation of the implementation of activities of the annual operations plan. For more detailssee chapter 5.

Examples of data held/processed in MIST include: ecological data, data about illegal activities,socio-economic data, tourism data, a literature database, an address database and annualoperations plans. MIST also gives users access to digital archives which consist of documents,spreadsheets and bitmaps stored on the server in UWA Headquarters. These documentsinclude quarterly and annual reports, minutes of meetings, forms, management plans, policydocuments, research findings, etc.

The UWA research project management database, a visual FoxPro application to catalogue andmonitor research projects and to assist during the approval process of new projects, can beaccessed through the MIST interface and will be integrated into MIST.

An overview of the components of MIST is given in figure 4.

5 An example of the use of financial data in MIST is the calculation of the costs for each square kilometrepatrolled by law-enforcement patrols. These costs are calculated based on ranger salaries and costs forfood rations and transport. MIST can then calculate the percentage of each PA’s budget spent on actuallypatrolling the area.

Standard monthly outputs produced by MIST for Murchison Falls Conservation Area

Maps patrol coverage (areas patrolled shown in 1 km² grid cells)

distribution and number of illegal activities

distribution and numbers of key wildlife species

Reports information on patrols (performance indicators such as time and distancepatrolled per ranger, costs per km patrolled, arrests per patrol day etc.)

illegal activities (indices and numbers)

key wildlife species (indices, numbers and population structure)


Page 14

Ranger Based Data CollectionMIST - GISAir patrolsCommunity Based Data CollectionAnnual Operations PlanningSystematic monitoring and census data

from the air and on the groundDigital Archives

Occasional observationsResearch dataAddress databaseTourism data

Literature database

Use-right database

Partner database

Research Project Management Database

Meta database

MIST Web browser

Fig. 4 The MIST toolbar and an overview of the MIST databases

4.4 Use of geo-referenced data for planning, decision-making, monitoring andevaluation

All spatial data collected are geo-referenced using Global Positioning System (GPS). The GPSused in UWA is the Garmin 12, a user-friendly, robust, hand-held unit with an in-built antenna.These units have been in use by ranger patrols since the end of 1997. The only problemsencountered to date have been occasional software failures6 which can be corrected atprotected area level using a Garmin software up-load function installed on the park computers.

The data collected by law-enforcement patrols in Uganda consist of:

• GPS waypoints (coordinates, date and time)• observations (mammals & numbers, sex and number of young animals, i.e. fit under

mother’s belly; illegal activities & numbers) and• names of the rangers on patrol.

Based on this small and easily collected set of geo-referenced raw data the following types ofoutputs are produced:

6 These failures are caused by simultaneously pressing a combination of buttons when starting the GPS.

Data collection by rangers on law-enforcement patrols must be simple, fast, designed tominimise data collection mistakes and provide raw data for a wide range of outputs.


Page 15

Indices

Indices provide measures of relative density and can be used in comparisons for monitoringwithout the need for expensive baseline data.

GPS readings by ranger patrols are taken at least every 30 minutes. This enables calculation ofthe distance covered by each patrol with reasonable accuracy, which in turn enables thecalculation of distance-related indices. Examples for distance-related indices are: elephantsseen per kilometre patrolled, snares collected per kilometre patrolled, costs per kilometrepatrolled (per month, per arrest etc.). An index such as snares collected per kilometre patrolledcan show changes in relative hunting pressure over time without knowing the actual number ofpoachers or number of animals killed. It can be used to indicate the success or failure of anti-poaching measures.

Ranger patrols do not systematically cover the entire park area. They are deployed based onlaw-enforcement requirements, areas patrolled recently, intelligence information etc. It istherefore essential to stratify (sub-divide) each protected area into strata (management sectors)which are similar in terms of patrolintensity.

Figure 5 shows the nine managementsectors of Murchison Falls ConservationArea. Areas patrolled during October 1998are displayed in 1 km² grid cells. An indexcalculated for the entire park based onsuch a deployment pattern would givemisleading results. Stratification of thepark into management sectors which havea high chance of being patrolled at asimilar intensity over time addresses theissue of patchy and unsystematic patroldeployment. Monitoring is done by lookingat indices (i.e. trends) for eachmanagement sector (e.g Buligi - light greyarea) over time.

Fig. 5 Management sectors of Murchison FallsConservation Area and areas patrolledduring October 1998

• Indices for monitoring of wildlife populations, illegal activities and resource harvest bylocal communities

• Distribution maps for planning and monitoring

• Baseline information for patrol deployment planning

• Information on wildlife population structure

• Performance indicators to monitor and evaluate the implementation of annual operationsplans

• All reporting requirements, e.g. the reports for MIKE (Monitoring of Illegal Killing ofElephants), a CITES initiative


Page 16

Indices can also be calculated from non geo-referenced data. Such indices are used incommunity based natural resource management to monitor resource off-take7 by localcommunities. Two indices are required to monitor the impact of the harvest on the resourcebase:

• the amount of resources harvested (e.g. headloads/month, handful/week) and• the effort required for the harvest of a defined quantity (e.g. hours/headload)

For example, if the amount harvested per month remains more or less constant over time onemight conclude that the harvest has no negative impact on the resource and that there isenough regeneration to sustainably support the current harvest volume. However, if at the sametime the effort to harvest a given amount (i.e. the time needed) increases significantly this mayindicate that the resource has been over-utilised close to the village and that people have to godeeper into the forest to meet their resource demands. This can indicate that naturalregeneration does not support the current harvest volume and that resource off-take is thereforeunsustainable.

Maps

Exact distribution maps can be used for monitoring and planning. Examples are:

Species distribution maps compiled from data collected over long periods will provideinformation about seasonal movement patterns and trends in distribution.

Maps showing the distribution of illegal activities are used for pro-active deployment of patrols.If, when looking at aggregated data over long periods, patterns become evident (e.g. increasedfishing activities in a certain area at the beginning of the dry season) this information can beused to deploy patrols pro-actively.

Patrol Deployment Planning

Deployment of patrols is done based on law-enforcement requirements and intelligenceinformation. In addition in MFCA, maps showing areas recently patrolled are used to prepare atwo week patrol operation plan.

Wildlife Population Structure

Information on wildlife population structure is used to assess the status of key wildlife species interms of sex ratio and number of young animals.

Performance Indicators

Performance indicators - such as number of patrol days per month and costs per kilometrepatrolled - are calculated as part of routine outputs and are used for monitoring and evaluationof the implementation of the annual operations plans. For more information see chapter 5.These indicators are normally related to the entire park but can also be related to amanagement sector (or any aggregated number of management sectors).

7 UWA allows communities living adjacent to protected areas to harvest certain resources which are notavailable outside the PA, such as medicinal plants and bamboo, from inside the protected area. Theharvest is done by authorised resource-users and the quota are fixed in a memorandum ofunderstanding.


Page 17

Reporting

Based on the geo-referenced data collected by ranger patrols MIST can also produce thereports required for MIKE (Monitoring Illegal Killing of Elephants8) without the need for anyadditional data collection.

4.5 Fast and accurate data entry

It is relatively easy to collect large amounts of data, but entering these data into a databasewithout mistakes is very time-consuming and risks creating a bottle-neck which makes itimpossible to provide up-to-date information. MIST avoids this problem through data downloadand the use of look-up lists for data entry. This not only avoids data entry errors, it also makesthe process of data entry easy and fast.

Most data collected at PA level are entered, imported and processed using the PC in theprotected area. Some data are entered by staff of the Information Management Unit (e.g. datacollected by researchers or data from large mammal counts from the air) either by hand orimported from GPS units and spread sheets.

An example of data recording, import, entry and export at PA level is given here using datacollected by ranger law-enforcement patrols.

Rangers record location data using GPS. Observations are recorded in one column of thedatasheet using an observation item and an observation code from a list in the observations boxat the bottom of the datasheet. The observation item describes the actual observation such aselephant (ELE) or sign of poaching (POS) while the observation code further qualifies theobservation item. Examples are sighting (no code), calls (c), dung (d) etc. for mammals andsnares (s), camps (c) etc. for signs of poaching. Any observations not listed in the observationsbox of the datasheet are recorded under remarks.

Waypoints, geographic coordinates, dates and times are down-loaded directly from the GPSinto MIST avoiding mistakes which would inevitably occur if data entry was done manually. Theyare displayed in an entry form which looks like the data sheet used by patrols. The only dataentry which has to be done is selection of observations (items and codes) from look-up tablesand entry of the number of observations. The names of the patrol members are also selectedfrom a look-up list which is linked to the UWA payroll database. It is important to note that thereis only one authorised version of each look-up table which is kept on the server and up-datesand changes can only be made by the system administrator at the IMU using the MISTDatabase Administrator Tool (see figure 8). These lists are then synchronised for all PAdatabases through database replication.

For illegal activities it is important to record more details than those provided by the observationcodes. These details include number of arrests, numbers and types of weapons confiscated etc.However, a datasheet which provides fields (columns) and tick boxes for the entry of suchdetailed recordings, which are only related to illegal activities, would become very large andthus unmanageable by law-enforcement patrols in the field. Rangers therefore record this levelof detail about illegal activities under remarks in a descriptive form. An overview of the datarecorded under remarks is given in table 1.

8 MIKE is a programme under CITES, the Convention on International Trade in Endangered Species ofWild Fauna and Flora.


Page 18

Table 1 Overview of data recorded by ranger patrolsObservationGroup

Observation Item Observation Code Observation remarks

MAMMALSBIRDSREPTILES

BuffaloElephantGorilla …

sightingdroppingfootprintcallnatural deathdead - cause unknownpoachednest (gorillas only)

cause of death: gun shot, snared, spear/arrow

action*weapons confiscated

Poaching[direct evidence]

huntingfishinghoney gathering

ILLEGALACTIVITIES

number arrested

number seen, not caught

animals confiscated

list of weapons seen

list of species

huntingfishinghoney gathering

Poaching Sign campswire snares/metal trapspoisonous baitsgun shots heardpeople (from foot prints)cartridgesfishing netscanoesbee hives

number arrestednumber seen, not caught

action*list of itemsconfiscated

Plant Harvesting[direct evidence]

pitsawingpole woodfire woodcharcoalbamboograssmedicinehandicraftspecimen collection

Plant HarvestingSign

pitsawingpole woodcharcoalbamboomedicine

trees felledpoles cutpitsstems cutnumber of plants

bark removedroots dug out

size of area (m²)number of hutsnumber of trespassersnumber of mines

number arrestedaction*

number seen, notcaught

cultivationsettlementtrespassingmininggrazing

Encroachment[direct evidence]

number of cattlenumber of sheepnumber of goats

EncroachmentSign- no people or livestockseen -

cultivationsettlementtrespassingmininggrazing

size of area (m²)number of hutsnumber of people (from foot prints)number of minesintensity (low, medium, high)

number arrestednumber seen, not caught

action*Water Collection[direct evidence]

Water CollectionSign

number of people

number (from foot prints)

Fire [direct evidence]Fire Sign

cautionedfined

Illegal touristtracking[direct evidence]

number of peopleamount USh

Off-road driving[direct evidence]

number of vehicles cautionedfined amount USh

* Actions: cautioned and released, fined (amount USh), detained at PA (number of days), taken to Local Councilhearing, taken to police (prosecuted)


Page 19

A box at the bottom of the datasheet lists the details which need to be recorded by the patrolsunder remarks. During data entry the relevant data need to be extracted from the writtenremarks. MIST provides a structured sequence of entry fields, look-up tables and tick boxeswhich guides the data entry person through this process. This enables the entry of very detaileddata without making the data sheet itself too complicated.

The same concept is applied to data entry for gorilla group monitoring. When visiting habituatedgorilla groups, rangers fill in a detailed gorilla group monitoring datasheet in addition to thepatrol datasheet. The data recorded on the gorilla group monitoring datasheet are entered aspart of the patrol observation data using a structured sequence of entry fields, look-up tablesand tick boxes in MIST.

4.6 Central information system (database maintenance and integrity)

The large amount of data to be processed in a national spatial information system requiresappropriate hardware and software. The system must also be user-friendly, easily accessibleand easy to maintain and up-date. A system which can only be used by a single, specialisedGIS (Geographic Information System) expert will inevitably collapse if the expert leaves theorganisation.

Criteria for the development of the IS and the selection of the software were:• A national IS must be in a position to handle large amounts of data in an efficient manner.• The data have to be stored in a central database for reasons of efficient maintenance and

data consistency and integrity.• The different types of information required by users need to be generated with a user

friendly front-end application which does not require the technical know-how of a GIS anddatabase expert.

• Users at all levels must have access to the IS from their workspace and several users mustbe able to access the IS simultaneously and independently.

• It is necessary to develop specific views on the application as well as on the databaseaccording to the different user groups that are going to have access to the system.

• All components of the IS (hardware and software) and the knowledge required to run andmaintain the IS have to be available or supported locally.

• The IS must provide for easy programme maintenance and updatability.

A major issue for database design is data integrity which refers to maintaining the correctnessand consistency of the data. Other important features are data independence (the data stored ina database are independent of the programme which uses them) and flexibility (it must bepossible to access data in many different ways and for many different purposes).

A database is a collection of structured data, the structure of which is independent of anyparticular application


Page 20

System Design

MIST has been developed as a client/server application to run in a Windows operating systemenvironment (NT Server and NT Workstations) and at the same time on stand alone PCs. Theprogramme consists of three logical units:

• a database and database server software• application processing libraries, and• a graphical user interface

The MIST programme libraries, which are the real-time processing system of the MIST userinterface, and the database server's software are both located on the server computer. Only theapplication interface is installed on each workstation (Fig. 6). All components of the applicationhave to be installed on a stand alone PC.

Borland Delphi was used to develop the user interface and libraries for MIST. Borland'sInterbase, version 6.0, SQL (Structured Query Language) database server handles datastorage and processing. Interbase consists of the SQL server and client software. Databasedata are stored on the server computer which uses the SQL server software to process adatabase request from a remote client (i.e. user's) computer and returns the appropriate data.

The SQL database containsattribute data (e.g. monitoring data)and other auxiliary data such asspecies look-up tables, databasesecurity files, etc. A central archive(master copy) of all spatial layers isalso located on the server. Fromthis central data archive, MIST canaccess and display all related andrelevant spatial layers linked to adatabase query9. The central dataarchive is considered an integraland necessary requirement ofMIST.

Fig. 6 The three logical units of MIST (database/server,libraries and interface)

9 A database query is the process of requesting data from the database and receiving the results.

A database server is a specialised process that manages the database itself. The clientapplications are workstation programmes which do not manipulate the database directly,they only make requests for the server to perform these operations. This allows the server toadd many sophisticated features, such as transaction processing, recovery, backup, accesscontrol etc. without increasing the size and complexity of the client application.

A central database ensuresconsistency and integrity of thedata from all protected areas andallows for efficient databasemaintenance.


Page 21

A user-friendly graphical interface is used to access the central data archive and createinformation requests against the archive. The programme interface is the only part of the overallsystem that most users will interact with. The programmes (see figure 4) that display the userinterface are only a small part of the many components of MIST. The purpose of these small,executable programmes is to connect the user to the database and to all necessary programmelibraries located on the server. These programmes do no direct data processing, this task is leftto the server’s database and libraries.

For example, the MIST-GIS module user interface displays spatial data stored in ESRI(Environmental Systems Research Institute) shape file format. The spatial display is handled byESRI Map Objects, which is completely integrated into MIST, and requires no external GISsoftware.

The visual aspect of the MIST-GIS user interface consists of two main components; these arethe Uganda national map and associated functions, and the local PA maps and associatedfunctions. Both of these main components have sub-functions designed to allow the user toaccess relevant information quickly.

Users can completely control what information they want from the programme by clicking onmap items, page tabs, list entries, and buttons. Most users will find the query wizards providedmore than adequate for their needs and will never need to use more advanced functions.However, more advanced users, or those with specific needs not directly supported by MIST-GIS, have increased flexibility with two additional tools: (1) a visual tool that allows the user torefine the SQL WHERE clause created by the general Query Wizard and (2) a visual SQLQuery Builder that permits users to create any number of new and unique queries. Therefore,MIST-GIS can be extended beyond its initial programming to adapt to the needs of the user inways that were either not envisaged during development or were not included in the programmebecause they are 'one-off' requests. All users, from users who are rarely exposed to computersto GIS specialists using MIST as a starting point for more sophisticated GIS analysis, should beable to generate the information they require from MIST.

The three levels for users to query the MIST database are:• easy-to-use Query Wizards (see figure 7)• a visual tool to refine the SQL WHERE clause created by the general Query Wizard and• a visual SQL Query Builder

Standardised reports are produced using the simple wizard shown in figure 7a for rangerreports. The user can only select the time period for a predefined output. A wide range offlexible outputs (maps, charts, tables and reports) are produced using the Query Wizard shownin figure 7b.

Maintenance of MIST does not require a GIS/database-specialist, it only requires a computerspecialist who can run a Windows NT-network and has a basic understanding of databasesoftware (e.g. can use InterBase or MIST to perform regular database back-ups, sweeps andsimple diagnostic procedures). UWA has got a computer network and a systems administratorto maintain the network. It therefore, does not require any additional, specifically trainedpersonnel to maintain the information system.

Most of the maintenance of the MIST database can be done using a visual MIST DatabaseAdministrator Tool (figure 8). This includes the definition and management of user privileges,updating/importing of monitoring data, updating of look-up tables, upkeep of high database


Page 22

B The user can select the:type of information

area to which the query applies (e.g. management sectors)time period (any period including disjunct periods)

observationstypes of output (differentkinds of maps, graphs and tables)

Fig. 7 MIST Query Wizards: A - for predefined outputs, B - for flexible output creation

A The user canselect the timeperiod for apredefined output

a result table isautomatically produced foreach query and can beexported in text delimited

format.


Page 23

performance, database back-ups and possible adjustments to the database procedures usingSQL.

Other important tasks for the computer specialist aremaintenance of the spatial data layers archive, regularvalidation of the database integrity, database replication usingIBReplicator and database back-ups on tape. All theseprocedures are documented in manuals which are kept in thedigital archives on the server. This ensures that knowledgedoes not get lost if the computer specialist leaves theorganisation, and that a new person can easily find and accessall information and procedures relevant for the smoothoperation of MIST and the computer network.

During system development importance was attached to thefact that the components of MIST should have a high degree oflocal support. ESRI's ArcView Shapefile format is widely usedby institutions in Uganda and is the default spatial format usedby MIST.

There was no need for MIST to be developed as an ArcViewreplacement, because digital spatial layers in ArcInfo Coverageor ArcView Shapefile formats are available from the NationalBiomass Study for the whole of Uganda. With these layersreadily available, UWA does not need to build capacity withinthe organisation to create spatial data layers, but instead canfocus on the use of spatial data in the information system.However, minor modifications/additions to existing shapefilessuch as new roads or new road alignments based on GPSreadings can be made by IMU staff using the functions providedby MIST.

Fig. 8 Database AdministratorTool in MIST

4.7 Access for all users

All users of MIST must have easy access to the central database from their workspace. Formanagers and planners at UWA Headquarters this is done through the existing NT local areanetwork. Members of the Board of Trustees can also use the local area network at UWAHeadquarters.

To enable access for the PAs and the Ministry, UWA has set up a wide area network which willhave two means of connection: (1) through the telephone network or (2) through wireless links.At present only direct dial-in through the mobile telephone network and the telephone land-lineshave been implemented. Data transfer through the mobile telephone network is limited to amaximum speed of 14.400 bps.

In future, wireless connections are planned for the Ministry and those PAs which have theirHeadquarters in towns which will be covered by the expansion of the wireless connectivities ofthe internet service provider used by UWA.

Protected areas without fast connection to the wide area network (i.e. without wireless links) willget access to MIST through stand alone computers in the PAs. The databases from these PAs


Page 24

Management

Protected Areas

Datacollection

PAdatabase

are synchronised with the central MIST database through database replication. Databasereplication is either done automatically using direct dial-in and digital data transfer or by usingzip-disks which are sent to UWA Headquarters for database replication.

An illustration of data flow from all levels, processing of data into information, communicationof the information to the users and access for managers and planners at all levels, to theinformation system is given in figure 9.

Fig. 9 Data and information flow and user access in the UWA Management Information System

All data entered at PA-level are replicated regularly to the central database at UWAHeadquarters. All changes made to the central database, such as up-dates of look-up tables,changes of the employee list etc., are replicated to the PA databases. This ensures integrity ofthe data and that only one authorised and validated version of the MIST database is in use.

The connection of PAs to MIST without wireless links has been made possible because Borlandrecently open sourced the Interbase database server. This open source version of Interbasecan be freely distributed which allows the installation of MIST on single PCs as a stand-alone

Data Management

Board of Trustees

Data Management

UWA Headquarters

Data Management

Ministry

Data provision. Data entry and data import (only protected areas and UWA HQ)

Database replication

Information for the management cycle - maps, graphics, tables, reports

Access to information at protected area level

Access to the information system (UWA computer network)

Access to the information system through direct dial-in and wireless link


Page 25

system without expensive site licenses. However, for consistency of the central databasearchitecture, no direct data entry or alterations on a local PC only system are allowed for mostdata. All this is done through database replication from the central database.

4.8 User-friendly programming contributes to sustainability

MIST’s user interface has been custom-made based on the requirements of the users in UWA.Such a user-friendly interface keeps the barrier to actually using MIST as low as possible. Asystem which is too complicated for use by managers (or which will not provide the requiredinformation), who do not have the time to learn complicated software, will not be used.Therefore the programming of a user-friendly interface for a custom-made information system isan important aspect contributing to the sustainability of MIST.

The second aspect contributing to sustainability has to do with the way the actual programmingis done and concerns those users who must maintain and eventually update MIST. Easyprogramme maintenance and programmability was considered as an important element in thedevelopment of MIST. This has taken two forms: the development of MIST as a group oflibraries that each deal with a single logical task, and the placement of much of the computingquery processes in control of the database server.

The central user interface of MIST is broken into several different Windows libraries that eachperform specific and encapsulated tasks for the user. This library system means that MISTupdates and re-development can progress at any time on any of these libraries in turn, on anas-needed basis without affecting the system as a whole.

The program libraries and SQL database procedures conduct all the business processing forMIST. They can both be adjusted or updated to alter how MIST functions. Depending on whatcomponent is updated, this can be used to change the user environment for specific users, toupdate or improve the computation performed, or to add new functionality to MIST.

By placing much of the code to run database queries directly into the database as SQLprocedures and triggers means that any minor, or even some substantial, changes can bemade to MIST without having to re-code the MIST user interface or the Windows libraries. Theability to alter existing libraries or to produce new ones makes MIST very updateable for futureneeds. The ability to add or alter MIST libraries, while more complicated than SQL procedurechanges, is more accessible, much simpler to perform and less costly than having to rewrite andadapt the MIST Windows programme for a new need that was not perceived at the time ofdevelopment.

The programming of MIST contributes to sustainability through:• a user-friendly interface• the use of programme libraries, each dealing with a single logical task, and• the possibility to up-date parts of the programme using SQL procedures and triggers


Page 26

5 MANAGEMENT EFFECTIVENESS

MIST is an efficient10 tool to improve management through easy access to up-to-dateinformation for decision-making. MIST is also used for planning and evaluation in themanagement cycle covering the six elements of evaluation (Table 2) defined by theManagement Effectiveness Task Force (World Commission on Protected Areas) to assess theeffectiveness of management of protected areas11.

Table 2 Framework for assessing the management effectiveness of protected areas andprotected area systems

Elements ofevaluation

Context Planning Input Process Output Outcome

Explanation where are wenow?

where do wewant to be?

what do weneed?

how do we goabout it?

what are theresults?

what have weachieved?

Assessmentof the

importance,threats andthe policyenvironment

PA design andplanning

resourcesneeded to carryout management

way in whichmanagementis conducted

quantity ofachievements

quality ofachievements

5.1 The management cycle

The management cycle involvesthe interaction of planning,implementation, monitoring,evaluation and replanning.

Uganda Wildlife Authority has putin place planning procedures forProtected Area GeneralManagement Plans and for AnnualOperations Plans. These plans arewritten within the framework set bythe UWA Strategic Plan. Therelationship between the differentlevels of planning is shown infigure 10.

Fig. 10 Hierarchy of planning in UWA

The general management planning process is an incremental process involving a series ofsteps from review through analysis of the current situation to the preparation of the plan. Inother words: deciding where you are now, making decisions about where you would like to beby a certain time, and creating realistic strategies for getting there.

10 Efficiency is a measure of the use of resources to achieve the desired results; effectiveness isproducing the desired/planned results.11 Hocking, M., Stolton, S. and Dudley, N. (2002). Evaluating effectiveness: a framework for assessing themanagement of protected areas. IUCN, Gland, Switzerland and Cambridge, UK. 121 pp.

UWAStrategic

Plan

5 years

Protected AreaGeneral Management

Plans

10 years

Annual Operations Plans

1 year


Page 27

UWA’s general management plan prescribes, for a 10 year period, the objectives and whatactions will be taken, why they will be taken and when they will be completed. Based on this,annual work plans or Annual Operations Plans (AOP) are developed. AOPs contain furtherdetails such as who is responsible; a schedule; a budget and indicators to evaluate theimplementation of the plan.

Planning is a linear process while the management system within which it operates is a cyclicalsystem involving the interaction of three key processes: Planning - Implementation -Evaluation.

There is little point in planning if progress after implementation is not monitored and evaluated.Evaluation examines implementation of the plan. Evaluation must be based on clearly definedperformance measures (indicators) which describe quality, quantity, place and time, and not onjust intuition, impressions or casual observations. Details of the annual management cycle areshown in figure 11.

Fig. 11 Annual management cycle

Planning and evaluation of the annual management cycle is handled in MIST.

• Data entry/import comprises: activity, person responsible for the implementation,indicator(s), means of verification, time frame, priority, budget, and whether anEnvironmental Scoping Brief (ESB) is required. The purpose of the ESB is to describeanticipated harm to the environment caused by the activity and what will be done tominimise the anticipated damage.

• Information for the quarterly evaluation of the status of implementation of the plan isprovided by MIST (e.g. calculation of indicators based on data collected by ranger patrols,gate clerks, local communities etc.) and by the accounting software used in UWA.

• Outputs such as quarterly and annual reports include detailed summary sheets for eachactivity, budget control reports, overview sheets with all activities, their time frame andimplementation status etc.

no

yes

AnnualOperations

PlanActivities Implementation

Monitoring&

Evaluation

deviationfrom plan

continue

steering

unplannedactivities

Quarterly planning cycle ( )

Annual planning cycle


Page 28

Assessment of the implementationstatus of activities is done using acode (A, B, C) according to theflow-chart shown in figure 12.These codes and explanations, ifrequired, are entered quarterly inMIST. X is used for activitieswhich have been postponed andV for activities which have beencompleted. If codes B or C areassigned an analysis of deviationhas to be entered in the database.For C in addition the correctiveaction has to be elaborated. Forpostponed activities (X) anexplanation has to be entered inMIST.

Fig. 12 Flowchart to illustrate how the evaluationassessment codes A – C are assigned

5.2 Measuring management effectiveness

Evaluation is a tool which helps managers to achieve the goals they set. It compares actualachievements with the targets set. It measures deviations from the plan and - if required -provides information for corrective actions to meet the targets set in the original plan, or tomodify the original plan. Evaluation ensures that implementation is done according to plan. It isnot a system for policing and punishment of inadequate performance - evaluation is a tool whichhelps management and, at the same time, ensures transparency.

Evaluation is neither a threat nor unnecessary extra work - it helps managers do their work andit provides the information to adjust future planning. For example, when using the indicators -number of patrol days per month, percentage of park area patrolled and costs per kilometrepatrolled - for law-enforcement patrols, management will get figures which not only prove theeffort they have put into patrolling the park, the figures will also prove whether or not man-powerand/or budget are sufficient to effectively and efficiently patrol the area. Any adjustment, ifnecessary, can then be made based on sound figures.

MIST improves management and measures management effectiveness through providing:

• Baseline data and information for planning• Information for evaluation• Indicators for institutional change

Baseline data and information for planning

MIST provides baseline data and can generate a wide range of spatial and tabular informationbased on:

• data collected by rangers, other UWA-staff, local communities, researchers etc., and• databases used/generated in MIST.

C

NO

YES

NO

YES

A

B

Evaluation ofimplementation of

activity

deviationfrom plan

need forcorrective

actions


Page 29

An example of databases used in MIST is the staff and salary database which is linked to theUWA payroll database. One of the databases generated in MIST is the budget database whichholds budget and expenditure figures from all annual operations plans.

In addition to information based on processed data, MIST also gives easy access to alldocuments etc. kept in the UWA digital archives.

Information for evaluation

The MIST-AOP module tells managers whether or not they have achieved what they planned.MIST provides the information for the evaluation and the MIST-AOP module provides the tool tomonitor and to record input in and output and outcome of management and to assessmanagement processes.

Evaluation (i.e. A-B-C coding and expenses) is done for each activity. All activities in MIST-AOPare linked to Units and Departments which in turn are grouped into cost centres12. Assessmentcan therefore be done at different levels from activity based costing to the assessment of theplan implementation at the organisational level (Table 3).

Table 3 The assessment levels in MIST-AOP

Evaluation based on

Evaluation levelsBudget/Expenses Implementation status (A-B-C)

Activity � �

Unit � �

Department � �

Cost centre � �

Organisation � �

Indicators of institutional change

Management is affected by human behaviour. It is therefore important to consider behaviouralaspects when introducing new management systems which will inevitably encounter oppositionfrom some individuals. Institutional changes are a dynamic process and initial opposition to newdevelopments will eventually be overcome and individuals will adjust.

MIST provides two indicators which measure how individuals adapt to new developments/institutional change through monitoring the changes in:

• the implementation status of activities, and• the use of MIST.

UWA recently introduced annual operations planning and managers need to learn how to planboth achievable activities and a realistic budget - they need to develop a planning culture. Thefirst annual planning cycle will inevitably produce a number of activities which cannot beachieved, which will have to be postponed and a number of unplanned activities, all indicatingan unrealistic plan. Once staff become more familiar with the annual planning cycle the

12 Each protected area and UWA headquarters are treated as separate cost centres.


Page 30

percentage of activities without deviations from the plan should increase, while at the same timethe number of activities which have to be postponed and the number of unplanned activities willdecrease. The relative changes of the percentage of A - C - X activities (for explanation seefigure 12) provide a simple indicator of how managers adapt to the introduction of annualoperations planning and evaluation.

Uganda Wildlife Authority has not only introduced a new management system, it has alsointroduced a new tool (MIST) to help managers implement the new system and to do their jobmore effectively and efficiently.

There has been a lot of initial opposition to using this new, computerised tool. Managers keepon asking their secretaries for information (which for example takes three days of compiling datafrom flat files and will inevitably contain numerous typographical errors) which they could get onbutton-click from MIST. When using MIST, managers will get the required information faster (i.e.more efficiently) and can make informed decisions (i.e. more effective management). The initialopposition will therefore be overcome once managers realise that by using MIST they can dotheir job more efficiently and effectively, prove what they have done and see that evaluation isnot a threat, but helps managing13.

The increase in the use of MIST is thus another indicator of how individuals adapt to a newmanagement system. The use of the various modules in MIST can be logged automatically byMIST.

13 The use of GPS units by law-enforcement patrols was met with a lot of initial opposition. However, thishas been overcome by now because the usefulness became obvious to managers in the PAs.

MIST is a custom-made, easy to use, flexible and powerful tool to improve managementefficiency and effectiveness. It is also a tool which can measures the effectiveness ofmanagement.


Page 31

6 META-DATABASE

Meta-data are data about data. They describe characteristics such as the content, quality andcondition of the data. The legend on a paper map, for example, can be considered as meta-data. For digital maps, documentation of meta-data is very important because producers andusers can modify the data or use them as source material for further products. Meta-data areessential when it comes to documentation, browsing and transfer of data.

In Uganda Wildlife Authority a lot of data exist in form of analogue and digital maps, databasetables and reports. To ensure optimal use of these data, they need to be inventoried anddocumented. The piling up of large amounts of undocumented data can render such datauseless because they cannot be found or understood. Only if data are properly documented in ameta-database can users find data they need for their intended application and assess theirsuitability regarding factors such as quality, cost or information content.

Imagine a situation in five years time when mainly new staff members need to assess theavailability of spatial baseline data on the road network of a PA for the revision of themanagement plan. Without proper meta-data documentation they could end up in a situationwhere they find numerous versions of the road map on different PCs and back-up media. Theywould not know which of the maps shows the correct road network. They would also not knowwhich parts of the road network are based on topographic maps, differential GPS readings,GPS readings before or after 01.05.200014 or on-screen digitisation based on field visits.

Data documentation contributes to institutionalising knowledge of an organisation’s data (seealso next section). The resignation of a member of staff could be a big loss if no writteninformation on data were available and nobody else knows about the data he was responsiblefor, what changes he made to spatial data and where master copies are kept.

Documentation of meta-data, ensuring data integrity and that duplication of work is avoided andproper storage of master copies are important tasks of the Information Management Unit.

UWA has decided to use the GRID meta-data directory to handle its meta-data. This meta-datatool follows international standards, thus providing the possibility for making information aboutUWA’s data available via the internet through the NASA’s Global Change Master Directory.

Data registration sheets for each dataset member have been designed in order to facilitatecollection of meta-data from internal or external data producers, to cover UWA’s more specificneeds for detailed data documentation and to ensure the collection of complete meta-data.

14 Around midnight of 01.05.2000 the United States stopped the intentional degradation of the GPSsignals from satellites. This increased the accuracy of GPS readings from about ±100 m to about ±5 m.


Page 32

knowledgeexchange

7 CAPACITY BUILDING THROUGH INFORMATION AND KNOWLEDGEMANAGEMENT

The project’s contribution to institutional and management capacity building also leads toknowledge management in UWA.

An overview of the relationship between the key components of the information system andknowledge management is shown in figure 13. The process from data capture through dataprocessing to the use of information is shown in blue (information management). Thedatabases, documents and the experience of individuals - the components of knowledgemanagement - are shown in black.

Fig. 13 Relationship between the key components of the information system and knowledgemanagement

Knowledge management is about (1) capturing and storing knowledge in an accessible way andabout (2) joining up people to enable knowledge flow/exchange across the organisation.

WORLD

Datacapture

Dataprocessing

Users

User process/understanding

Decisions,actions etc.

Central database

Meta-database

User's memory,knowledge base

REAL

Effectsof actionsanddecisions

Access Dissemination

other inputs:external data,documents,manuals,etc.

other inputs:informal IS,training,etc.

Information

Knowledge management is the broad process of locating, organising, transferring and usingthe information and expertise within an organisation. This includes databases and documentsas well as the sum of the experience of all individuals in the organisation.


Page 33

Knowledge management uses databases and documents to capture information within anorganisation. MIST gives easy access to several databases (see figure 4) and the UWA digitalarchives.

Another important aspect of knowledge management is the experience of individuals and howto transfer personal into institutional knowledge. Such knowledge can either be communicatedand shared easily (for example procedures and instructions) or it can be personal knowledgewhich is more difficult to identify, formalise and communicate.

Procedures carried out by individuals – such as how to record meta-data, how to set-up a GPS,how to maintain the MIST database etc. – are documented in manuals in such a way that theknowledge does not get lost if a particular person leaves the organisation. These manuals areeasily accessible through the digital archives and MIST.

Personal knowledge concerns people and can only be communicated if the culture andstructure of the organisation enables them to do this. It is therefore important to create a cultureof knowledge exchange and establish this as a core value of the organisation if knowledgemanagement is to be achieved.

MIST contributes to institutionalising information and knowledge management by providingall staff in UWA with easy access to the central database, the meta-database and alldocuments and manuals of relevance to Uganda Wildlife Authority. It also contributes tocreating a culture of knowledge exchange.

Documents

INFORMATION AND KNOWLEDGE MANAGEMENT IN NATURE CONSERVATION